mlsanityData to 14 Oct 2026

SGFormer: Satellite-Ground Fusion for 3D Semantic Scene Completion

arXiv 2503.16825 ↗First posted 21 Mar 2025cs.CV cs.ROcited by 3

Not on the map. Why?

Abstract (arXiv)

Recently, camera-based solutions have been extensively explored for scene semantic completion (SSC). Despite their success in visible areas, existing methods struggle to capture complete scene semantics due to frequent visual occlusions. To address this limitation, this paper presents the first satellite-ground cooperative SSC framework, i.e., SGFormer, exploring the potential of satellite-ground image pairs in the SSC task. Specifically, we propose a dual-branch architecture that encodes orthogonal satellite and ground views in parallel, unifying them into a common domain. Additionally, we design a ground-view guidance strategy that corrects satellite image biases during feature encoding, addressing misalignment between satellite and ground views. Moreover, we develop an adaptive weighting strategy that balances contributions from satellite and ground views. Experiments demonstrate that SGFormer outperforms the state of the art on SemanticKITTI and SSCBench-KITTI-360 datasets. Our code is available on https://github.com/gxytcrc/SGFormer.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 34 of its 45 references that are in the map. First posted on arXiv 21 Mar 2025.

Builds on

34 of 45 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  3. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  4. 2017Squeeze-and-Excitation Networkscited by 4,789
  5. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  6. 2018CBAM: Convolutional Block Attention Modulecited by 2,477
  7. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,218
  8. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,044
  9. 2016Semantic Scene Completion from a Single Depth Imagecited by 472
  10. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 409
  11. 2021MonoScene: Monocular 3D Semantic Scene Completioncited by 363
  12. 2023VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completioncited by 313
22 more
  1. 2023OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Predictioncited by 293
  2. 2020Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completioncited by 223
  3. 2020LMSCNet: Lightweight Multiscale 3D Semantic Completioncited by 188
  4. 2019Lending Orientation to Neural Networks for Cross-view Geo-localizationcited by 180
  5. 2024GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Predictioncited by 157
  6. 2020S3CNet: A Sparse Semantic Scene Completion Network for LiDAR Point Cloudscited by 102
  7. 2024SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Predictioncited by 96
  8. 2023Symphonize 3D Semantic Scene Completion with Contextual Instance Queriescited by 90
  9. 2022Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Imagecited by 88
  10. 2023OccDepth: A Depth-Aware Method for 3D Semantic Scene Completioncited by 86
  11. 2023OrienterNet: Visual Localization in 2D Public Maps with Neural Matchingcited by 84
  12. 2021MobileStereoNet: Towards Lightweight Deep Networks for Stereo Matchingcited by 82
  13. 2023SCPNet: Semantic Scene Completion on Point Cloudcited by 80
  14. 2023SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Drivingcited by 75
  15. 2022Uncertainty-aware Vision-based Metric Cross-view Geolocalizationcited by 54
  16. 2024Context and Geometry Aware Voxel Transformer for Semantic Scene Completioncited by 50
  17. 2023SNAP: Self-Supervised Neural Maps for Visual Positioning and Semantic Understandingcited by 38
  18. 2023S4C: Self-Supervised Semantic Scene Completion with Neural Fieldscited by 28
  19. 2024GEOcc: Geometrically Enhanced 3D Occupancy Network with Implicit-Explicit Depth Fusion and Contextual Self-Supervisioncited by 21
  20. 2024SG-BEV: Satellite-Guided BEV Fusion for Cross-View Semantic Segmentationcited by 18
  21. 2022Any Way You Look At It: Semantic Crossview Localization and Mapping with LiDARcited by 12
  22. 2024Increasing SLAM Pose Accuracy by Ground-to-Satellite Image Registrationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025VoxDet: Rethinking 3D Semantic Occupancy Prediction as Dense Object Detectioncited by 13
  2. 2025OccuFly: A 3D Vision Benchmark for Semantic Scene Completion from the Aerial Perspectivecited by 5
  3. 2026SparseOcc++: Geometry-Aware Sparse Latent Representation for Semantic Occupancy Predictioncited by 0